CN120066057B - Multi-robot collaborative handling method and system based on multi-objective optimization irregular object - Google Patents
Multi-robot collaborative handling method and system based on multi-objective optimization irregular object Download PDFInfo
- Publication number
- CN120066057B CN120066057B CN202510550205.5A CN202510550205A CN120066057B CN 120066057 B CN120066057 B CN 120066057B CN 202510550205 A CN202510550205 A CN 202510550205A CN 120066057 B CN120066057 B CN 120066057B
- Authority
- CN
- China
- Prior art keywords
- robot
- coordinates
- irregular
- objective
- robots
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/40—Control within particular dimensions
- G05D1/43—Control of position or course in two dimensions
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/60—Intended control result
- G05D1/644—Optimisation of travel parameters, e.g. of energy consumption, journey time or distance
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/60—Intended control result
- G05D1/69—Coordinated control of the position or course of two or more vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a multi-robot collaborative handling method and a multi-target optimization-based irregular object multi-robot collaborative handling system, which belong to the technical field of multi-mobile robot collaborative control and comprise the steps of obtaining shape information of an object and extracting edge contour coordinates of the object; the method comprises the steps of converting edge contour coordinates of an object into parameter equations of actual action points at the tail end of a connecting rod of a robot, expanding the shape of the object, establishing a mapping relation between the parameter equations and the reachable positions to obtain a set of actually reachable position points of the robot, constructing a multi-objective optimization function, optimizing coordinate points serving as variables to be optimized by using an improved NSGA-II algorithm to obtain optimal solutions, determining optimal action point coordinates according to the number of the robots, and controlling multiple robots to carry out rigid formation to complete cooperative carrying tasks of irregular objects after the optimal action point coordinates are obtained. The invention solves the problems of object shape analysis missing, formation constraint shortage and multi-objective collaborative optimization in the prior art, thereby realizing the high efficiency and stability of the carrying task.
Description
Technical Field
The invention belongs to the technical field of multi-mobile robot cooperative control, and particularly relates to a multi-robot cooperative conveying method and system for irregular objects based on multi-objective optimization.
Background
The statements herein merely provide background information related to the present disclosure and may not necessarily constitute prior art.
With the continuous development of robot technology in recent years, a multi-robot system becomes a research hotspot in the fields of global robotics and control. Collaborative handling is one of the popular research applications of multi-robot systems, which aims at moving a target object to a specific destination by a group of robots, and has an advantage in that a single robot cannot independently handle an object having excessive quality. Many factors need to be considered in designing and implementing the collaborative handling system, such as what multi-robot communication mechanism, coordination algorithm, control architecture, task allocation mechanism, handling platform, etc. are selected.
The multi-robot collaborative handling technology is a research hot spot due to the expandability and high efficiency, and the existing mainstream handling strategies include a push-only strategy (pushing objects by robots), a caging strategy (enclosing objects in a formation center), and a grabbing strategy (grabbing objects with high-degree-of-freedom mechanical arms). However, the inventor discovers that the strategies have the defects that the geometrical characteristics of irregular objects are not accurately modeled due to the lack of modeling basis for the selection of action points, the suitability of robot formation with the shapes of the objects is not considered due to the lack of rigidity constraint, the stability is insufficient or the efficiency is low in the carrying process easily, the multi-objective collaborative optimization is blank, and the conventional method does not comprehensively balance key indexes such as the energy consumption, the carrying efficiency and the safety of the system.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a multi-robot collaborative handling method and a system based on multi-objective optimization irregular objects, which are characterized in that the object shape is accurately modeled, the action position of the multi-robot is optimized, a multi-objective optimization function is constructed, an improved NSGA-II algorithm is utilized to obtain an optimal solution, a DMPC controller with rigid formation constraint is designed to complete collaborative handling tasks, and the problems of object shape analysis deficiency, insufficient formation constraint and multi-objective collaborative optimization in the prior art are solved, so that the high efficiency and stability of the handling tasks are realized.
In order to achieve the above object, the present invention is realized by the following technical scheme:
on one hand, the technical scheme of the invention provides a multi-robot collaborative handling method based on multi-objective optimization of irregular objects, which comprises the following steps:
acquiring shape information of an object, and extracting edge contour coordinates of the object;
Converting the edge contour coordinates of the object into a parameter equation of an actual acting point at the tail end of the robot connecting rod;
Expanding the shape of an object, and establishing a mapping relation between a parameter equation and an reachable position to obtain a practical reachable position point set of the robot;
Constructing a multi-objective optimization function, taking coordinate points as variables to be optimized, optimizing initial population quality by using an improved NSGA-II algorithm, defining constraint violation degrees, further solving an optimal solution, and determining the coordinates of optimal action points according to the number of robots;
and after the optimal action point coordinates are obtained, controlling the multiple robots to carry out rigid formation to complete the cooperative carrying task of the irregular object.
In at least one embodiment, the method comprises the steps of obtaining the shape information of the object and extracting the edge contour coordinates of the object, specifically, obtaining an original image of the object through an RGB camera, converting the original image into a gray scale image, performing Gaussian filtering, and then extracting the edge contour coordinates of the object through an edge detection algorithm.
In at least one embodiment, the method for converting the edge profile coordinates of the object into the parameter equation of the actual action point of the tail end of the robot connecting rod specifically comprises the step of converting the edge coordinates of the object into the parameter equation of the actual action point of the tail end of the robot connecting rod by adopting a corresponding interpolation method according to the type of the object.
In at least one embodiment, the object types include closed polygons and smooth closed curves.
In at least one embodiment, the parametric equation is constructed using piecewise linear interpolation if the object type is a closed polygon, and cubic spline interpolation if the object type is a smooth closed curve.
In at least one embodiment, expanding the shape of an object, establishing a mapping relation between a parameter equation and an reachable position to obtain an actual reachable position point set of the robot, wherein the expanding length is the length of a connecting rod of the robot according to the shape of the object described by the parameter equation and is outwards expanded along the normal direction, the mapping relation between the parameter equation and the reachable position is established to obtain a two-dimensional coordinate point to be selected which can be directly reached by the robot, invalid points caused by non-convexity or volume conflict of the object are removed, and finally the actual reachable position point set after removal is obtained.
In at least one embodiment, the constructing the multi-objective optimization function is specifically:
The method comprises the steps of constructing a multi-objective optimization function according to the corresponding position of each robot and the position corresponding to the mass center of an object, maximizing the distance between robots, constructing a position dispersion objective function, constructing a moment maximization objective function based on enabling each robot to be far away from the center of an irregular object as far as possible, constructing a shape closure objective function based on the symmetric distribution of acting force, and adding collision constraint among the robots to enable the distance between the robots to meet a safe distance threshold.
In at least one embodiment, based on the constructed multi-objective function and collision constraint, using an improved NSGA-II algorithm to take an object parameter equation and the number of robots as input, iterating out the optimal solution of each objective function as an individual in an initial population, then iterating out the solution to obtain the pareto front, adjusting the weights of the position-dispersed objective function, the moment-maximized objective function and the shape-closed objective function at the pareto front according to actual task requirements, and determining the optimal solution through weighted summation to obtain the optimal reachable position coordinates and the optimal action point coordinate information of the multi-robots.
In at least one embodiment, after the coordinates of the optimal action points are obtained, the tail ends of the connecting rods of the robots are rigidly connected to the object, the formation pose is adjusted in real time through a controller with rigid formation hard constraints, and the robots are controlled to complete the cooperative conveying task of the irregular object.
On the other hand, the technical scheme of the invention also provides a multi-robot collaborative handling system based on multi-objective optimization irregular objects, which comprises the following steps:
The edge contour coordinate extraction module is configured to acquire shape information of an object and extract edge contour coordinates of the object;
the parameter equation construction module is configured to convert edge contour coordinates of an object into a parameter equation of an actual action point of the tail end of the robot connecting rod;
The reachable position point set generating module is configured to expand the shape of the object, establish a mapping relation between a parameter equation and the reachable position and obtain an actual reachable position point set of the robot;
The multi-objective optimization module is configured to construct a multi-objective optimization function, take coordinate points as variables to be optimized, optimize the initial population quality by utilizing an improved NSGA-II algorithm, define constraint violation degrees, further obtain an optimal solution, and determine the coordinates of the optimal action points according to the number of robots;
and the task execution module is configured to control the multiple robots to carry out rigid formation to complete the cooperative carrying task of the irregular object after the optimal action point coordinates are obtained.
The technical scheme of the invention has the following beneficial effects:
1) According to the multi-robot collaborative handling method and system based on the multi-objective optimization irregular object, the object shape is accurately modeled, the action positions of the multi-robot are optimized, the multi-objective optimization function is constructed, the improved NSGA-II algorithm is utilized to obtain the optimal solution, and the DMPC controller with the rigid formation constraint is designed to complete the collaborative handling task, so that the problems of object shape analysis deficiency, insufficient formation constraint and multi-objective collaborative optimization in the prior art are solved, and the high efficiency and stability of the handling task are realized.
2) The multi-robot collaborative handling method and system based on the multi-objective optimization irregular object can enable shape modeling to have high efficiency, the irregular object contour data is compressed and stored through a parameter equation, the calculation complexity is reduced, the relationship between the solution of the optimization problem and the actual mobile robot position and the action point position is clearly established by utilizing the expansion of the irregular object shape, and the subsequent expansion of the position distribution and track control module is facilitated.
3) The multi-robot collaborative handling method and system based on the multi-objective optimization irregular object can realize multi-objective collaborative optimization, comprehensively balance the stability, efficiency and safety of the whole collaborative handling system, and promote the adaptability of the system.
4) The multi-robot collaborative handling method and system based on the multi-objective optimization irregular object can enhance the practicability of an optimization algorithm, remarkably shorten the solving time, can be suitable for multi-AGV collaborative handling of large irregular objects in an actual industrial scene, and has wide engineering application value.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a schematic overall flow diagram of the multi-robot collaborative handling method based on multi-objective optimization of irregular objects of the present invention.
Fig. 2 is a schematic view of object contour extraction and action position optimization according to the present invention, wherein (a) is a schematic view of an edge contour map and a selected action point of a rectangular-like object, (b) is a schematic view of an edge contour map and a selected action point of an L-like object, and (c) is a schematic view of an edge contour map and a selected action point of an elliptical-like object.
Detailed Description
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
As described in the background art, the present invention aims to overcome the shortcomings of the prior art, and provide a multi-robot collaborative handling method and system based on multi-objective optimization of irregular objects, which can optimize the positions of the multi-robots by accurately modeling the shapes of the objects, the multi-objective optimization function is constructed, an improved NSGA-II algorithm is utilized to obtain an optimal solution, a DMPC controller with rigid formation constraint is designed to complete a collaborative handling task, and the problems of object shape analysis deficiency, insufficient formation constraint and multi-objective collaborative optimization in the prior art are solved, so that the high efficiency and the stability of the handling task are realized.
Example 1
In an exemplary implementation manner of the present invention, as shown in fig. 1, the present embodiment discloses a multi-robot collaborative handling method based on multi-objective optimization of irregular objects, including:
s100, acquiring shape information of an object, and extracting edge contour coordinates of the object;
S200, converting the edge contour coordinates of the object into a parameter equation of an actual action point of the tail end of the robot connecting rod;
S300, expanding the shape of an object, generating a set of reachable position points of the robot, and establishing a mapping relation between a parameter equation and reachable positions;
S400, constructing a multi-objective optimization function, taking coordinate points as variables to be optimized, optimizing initial population quality by utilizing an improved NSGA-II algorithm, defining constraint violation degrees, further solving an optimal solution, and determining the coordinates of the optimal action points according to the number of robots;
and S500, after the optimal action point coordinates are obtained, controlling the multiple robots to carry out rigid formation to complete the cooperative transportation task of the irregular object.
The multi-objective optimization-based irregular object multi-robot collaborative handling method is described in detail below with reference to specific embodiments.
S100, acquiring shape information of the object, and extracting edge contour coordinates of the object.
S101, acquiring images, namely erecting an RGB industrial camera with the resolution of 1920 x 1080, and acquiring a top view image of an irregular object to be used as an original image of the irregular object.
S102, gray level and denoising of the image, namely carrying out gray level treatment on the original image to obtain a gray level image of an irregular object, and carrying out Gaussian filtering treatment on the gray level image to eliminate interference of noise on edge detection. The conversion mode of graying specifically comprises the following steps:
。
S103, extracting edge contour coordinates, namely extracting the edge contour coordinates of the irregular object from the gray level image subjected to denoising treatment by utilizing a Canny edge detection algorithm through steps of image gradient calculation, non-maximum value suppression, double-threshold edge determination and the like.
Specifically, the edge profile is mainly based on the characteristic that the gray level of the image changes drastically. The Canny algorithm mainly comprises the following steps:
(1) Calculating the gradient of the image, and calculating the gradient and the direction of each pixel by using a Sobel operator;
(2) Suppressing non-maximum value, calculating whether the gradient amplitude of the current pixel point is a local maximum value along the gradient direction, and only reserving the pixel point with the local maximum gradient value as an edge to suppress other points;
(3) The double threshold value determines the edge, and the non-edge points generated by the self band or noise of the image are filtered through the high threshold value and the low threshold value.
And S200, converting the edge contour coordinates of the object into a parameter equation of an actual acting point of the tail end of the connecting rod of the robot.
Based on the shape information of the irregular object obtained in S100 and the extracted edge contour coordinates, the types of the irregular object can be divided into closed polygons and smooth closed curves, and the edge contour coordinates of the shape of the irregular object are converted into a parameter equation by a piecewise linear interpolation method or a cubic spline interpolation method according to the types of the objectThe method is used for describing contour information of an irregular object, namely an actual action point of the tail end of the connecting rod of the mobile robot, and setting a range of a parameter t to be t=0, wherein t=0 corresponds to a starting point, and t=1 corresponds to an ending point, namely a return to the starting point.
Specifically, S201 if the object type is a closed polygon, the interval is defined byN equal division is carried out, each subinterval corresponds to a line segment between two adjacent points, and an edge coordinate value is calculated in each subinterval in a linear interpolation mode. The specific form of the parameter equation is as follows:
for the ith line segment When (when)At the time, there are
,
,
Wherein the method comprises the steps ofAndRepresenting the x-coordinate and y-coordinate of the ith corner point constituting the closed polygon whenIn the time-course of which the first and second contact surfaces,To ensure closure.
S202, if the object is not a closed polygon, but a smooth closed curve, adopting cubic spline interpolation to represent, and constructing a piecewise cubic polynomial to ensure that the first derivative and the second derivative of adjacent segments at the connecting points are continuous, thereby forming an overall smooth curve. In parameter allocation, the allocation is accumulated according to the distance between adjacent points,For each intervalThe corresponding cubic polynomial is as follows:
,
,
wherein the coefficient is Determined by solving the following conditions:
(a) Continuous position ;
(B) First derivative continuous;
(C) Second derivative continuous;
(D) Periodic boundary conditions。
S300, expanding the shape of the object, generating a set of reachable position points of the robot, and establishing a mapping relation between a parameter equation and the reachable position.
Specifically, the object shape described by the parameter equation expands outwards along the normal direction, the expansion length is the length L of the robot connecting rod, and the unit normal vector of the object shape isEstablishing a mapping relation between a parameter equation and an reachable position as followsAnd obtaining a two-dimensional coordinate point to be selected, which can be directly reached by the robot.
In the actual expansion processing, because the non-convexity of the object shape can lead expansion points not to be reached in the object or after the robot volume is considered, the points need to be removed, the non-smooth points in the object shape parameter equation can not directly exert an effect in practice, the points do not have expansion points, the situation that two or more edge points correspond to the same expansion point can be generated due to the particularity of the object shape, and the effect of applying the two or more edge points after the mobile robot reaches the expansion point is considered to be the same for simplifying the processing, and finally, the practically reachable position point set S after removal is obtained.
S400, constructing a multi-objective optimization function, taking coordinate points as variables to be optimized, optimizing initial population quality by using an improved NSGA-II algorithm, defining constraint violation degrees, further solving an optimal solution, and determining the coordinates of the optimal action points according to the number N of robots. The specific steps are as follows:
S401, firstly constructing a multi-objective function, specifically:
(1) According to the corresponding position of each robot The corresponding position of the mass center of the object isThe multi-objective optimization function is constructed as follows:
Wherein, the 。
(2) Position dispersion targets, namely maximizing the distance between robots, improving the stability of a multi-mobile robot collaborative handling system, and establishing a position dispersion target function as。
(3) Moment maximizing goal, namely enabling each mobile robot to be far away from the mass center of an irregular object as far as possible (assuming that the mass of the object is uniformly distributed), improving the moment output upper limit, and establishing a moment maximizing objective function as。
(4) The object of shape closure is to ensure symmetric distribution of acting force, and to establish the shape closure object function as。
(5) Adding collision constraint between mobile robots, the distance between robots needs to meet a safe distance threshold,Wherein R is the length of the connecting rod,Is a safe elastic distance.
S402, optimizing initial population quality by using an improved NSGA-II algorithm, defining constraint violation degree, and further obtaining an optimal solution.
Based on the constructed multi-objective function and collision constraint, the pareto front corresponding to the solving optimization problem in the NSGA-II algorithm is improved.
In this embodiment, the initial population is generated by real number encoding with the path parameters in the parametric equation as decision variables. Initializing the size of an input population, the cross probability and the variation probability in the population, and solving the optimal solution of the single objective function, wherein the optimal solution is used as a high-quality initial individual in the initial population so as to reduce the iteration times and accelerate the solving efficiency.
For constraint problems, constraint violations are defined,. Defining dominance relations in non-dominance ordering, feasible solutions in hierarchical comparisonsAlways is superior to the infeasible solutionThe infeasible solutions are incrementally ordered by value. If two solutions are adoptedAndAre all feasible solutions, andNot inferior to all targetsAnd is strictly better at least on one target, thenDominance ofIf both solutions are not feasible,Smaller solutions are more optimal, layering is carried out according to the sorting rule and the pareto dominant relationship, non-dominant solutions are classified into a first layer Front1, the rest after the non-dominant solutions are removed is classified into a second layer Front2, and the like. Only considering the distribution of feasible solutions when calculating the degree of congestion, ensuring the diversity of the feasible solutions, and preferentially eliminating when merging the parent and the offspringIs a solution to (a).
Iteration is carried out to obtain a feasible pareto front, all solutions meet the constraint condition of an objective function, finally, position dispersion, moment maximization and shape sealing objective function weight are given according to actual task requirements, and the optimal solution is determined through weighted summationNamely, the best reachable position coordinates and the best action point coordinate information of N robots.
Solving in an MATLAB platform by designing the multi-objective function and constraint, combining single-objective optimization pre-solving in population initialization optimization to generate a high-quality initial individual, setting the population scale to be 50, the crossover probability to be 0.7 and the variation probability to be 0.4, firstly solving the optimal solution of the single-objective function through 3 times of 200 iterations to serve as the high-quality individual of the multi-objective optimization initial population, and then screening the pareto front solution through 200 times of iterations. And (3) distributing weights of indexes such as position dispersity, moment upper limit, shape sealing performance and the like based on actual task demands, and determining final deployment positions of the N robots through weighted summation. As shown in fig. 2, the edge extraction and the action point optimization results of different types of objects are tested on three types of typical target forms (the rectangular-like object in fig. 2 (a), the L-shaped object in fig. 2 (b) and the elliptical-like object in fig. 2 (c) can be respectively regarded as a regular object, a regular polygonal object and an irregular object), and the improved algorithm can be converged to an optimized position configuration meeting geometric constraint with high efficiency.
And S500, after the optimal action point coordinates are obtained, controlling the multiple robots to carry out rigid formation to complete the cooperative transportation task of the irregular object.
After the coordinates of the optimal action points are obtained, a Mecanum wheel mobile robot with a passive single-degree-of-freedom rotary connecting rod is designed, the tail end of the connecting rod is rigidly connected to an object, track tracking is realized through a DMPC controller with rigid formation hard constraint, formation postures are adjusted in real time, the mobile robot is controlled to complete the cooperative carrying task of an irregular object, and the dynamic stability of a system in the carrying process is ensured. The results show that the system shows high robustness and effectiveness in complex shape object handling.
The method comprises the steps of obtaining an original image through an erected RGB camera, obtaining contour coordinate point set information of an irregular object by utilizing an image processing technology, obtaining a mapping relation between parameters and contour coordinate points by establishing a parameter equation, expanding the object contour outwards along a normal direction according to the length of a connecting rod in a designed mobile robot carrying mechanism, obtaining a coordinate point set which can be acted and reached by the mobile robot, obtaining a mapping relation between the parameters and reachable coordinate points, designing a multi-objective optimization function considering carrying safety and stability, a carrying upper limit and efficiency, taking the object parameter equation and the number of robots as input by utilizing an improved NSGA-II algorithm, iterating out an optimal solution of each objective function as an individual in an initial population, iterating the solution to obtain a pareto front, adjusting weights of three-item objective function at the pareto front according to the requirement in actual carrying, and obtaining the optimal reachable position coordinates and optimal action point coordinate information of the multi-robot.
Example 2
In an exemplary implementation manner of the present invention, the present embodiment discloses a multi-robot collaborative handling system based on multi-objective optimization of irregular objects, including:
The edge contour coordinate extraction module is configured to acquire shape information of an object and extract edge contour coordinates of the object;
the parameter equation construction module is configured to convert edge contour coordinates of an object into a parameter equation of an actual action point of the tail end of the robot connecting rod;
the reachable position point set generating module is configured to expand the shape of the object, generate a reachable position point set of the robot and establish a mapping relation between a parameter equation and the reachable position;
The multi-objective optimization module is configured to construct a multi-objective optimization function, take coordinate points as variables to be optimized, optimize the initial population quality by utilizing an improved NSGA-II algorithm, define constraint violation degrees, further obtain an optimal solution, and determine the coordinates of the optimal action points according to the number of robots;
and the task execution module is configured to control the multiple robots to carry out rigid formation to complete the cooperative carrying task of the irregular object after the optimal action point coordinates are obtained.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The multi-robot collaborative handling method based on multi-objective optimization irregular object is characterized by comprising the following steps:
acquiring shape information of an object, and extracting edge contour coordinates of the object;
Converting the edge contour coordinates of the object into a parameter equation of an actual acting point at the tail end of the robot connecting rod;
Expanding the shape of an object, and establishing a mapping relation between a parameter equation and an reachable position to obtain a practical reachable position point set of the robot;
Constructing a multi-objective optimization function, taking coordinate points as variables to be optimized, optimizing initial population quality by using an improved NSGA-II algorithm, defining constraint violation degrees, further solving an optimal solution, and determining the coordinates of optimal action points according to the number of robots;
and after the optimal action point coordinates are obtained, controlling the multiple robots to carry out rigid formation to complete the cooperative carrying task of the irregular object.
2. The multi-robot collaborative handling method based on multi-objective optimization irregular objects according to claim 1, wherein the method is characterized in that the shape information of the objects is obtained, and the edge contour coordinates of the objects are extracted, specifically, the original images of the objects are obtained through an RGB camera, the original images are converted into gray-scale images and are subjected to Gaussian filtering, and then the edge contour coordinates of the objects are extracted through an edge detection algorithm.
3. The multi-objective optimization irregular object-based multi-robot collaborative handling method according to claim 1, wherein the method is characterized in that the edge contour coordinates of the object are converted into parameter equations of actual action points of the tail ends of the robot connecting rods, specifically, according to the type of the object, the edge coordinates of the object are converted into parameter equations of the actual action points of the tail ends of the robot connecting rods by adopting a corresponding interpolation method.
4. A multi-objective optimized irregular object based multi-robot co-handling method according to claim 3, wherein the object types include closed polygons and smooth closed curves.
5. The multi-objective optimization irregular object-based multi-robot collaborative handling method according to claim 4, wherein if the object type is a closed polygon, a piecewise linear interpolation method is adopted to construct a parameter equation, and if the object type is a smooth closed curve, a cubic spline interpolation method is adopted to construct the parameter equation.
6. The multi-objective optimization irregular object-based multi-robot collaborative handling method is characterized by comprising the steps of expanding object shapes, establishing a mapping relation between a parameter equation and an reachable position to obtain an actual reachable position point set of a robot, specifically, expanding the object shapes described by the parameter equation outwards along a normal direction, wherein the expansion length is the length of a connecting rod of the robot, establishing a mapping relation between the parameter equation and the reachable position to obtain a two-dimensional coordinate point to be selected, which can be directly reached by the robot, eliminating invalid points caused by non-convexity or volume conflict of the object, and finally obtaining the actual reachable position point set after elimination.
7. The multi-objective optimization-based irregular object multi-robot collaborative handling method according to claim 1, wherein the constructing a multi-objective optimization function specifically comprises:
The method comprises the steps of constructing a multi-objective optimization function according to the corresponding position of each robot and the position corresponding to the mass center of an object, maximizing the distance between robots, constructing a position dispersion objective function, constructing a moment maximization objective function based on enabling each robot to be far away from the center of an irregular object as far as possible, constructing a shape closure objective function based on the symmetric distribution of acting force, and adding collision constraint among the robots to enable the distance between the robots to meet a safe distance threshold.
8. The multi-objective optimization irregular object multi-robot collaborative handling method according to claim 7, wherein based on constructed multi-objective functions and collision constraints, an object parameter equation and the number of robots are used as input by using an improved NSGA-II algorithm, an optimal solution of each objective function is firstly iterated out to serve as an individual in an initial population, a pareto front is obtained by means of iterative solution, weights of a position dispersion objective function, a moment maximization objective function and a shape sealing objective function are adjusted at the pareto front according to actual task requirements, and an optimal solution is determined through weighted summation, so that optimal reachable position coordinates and optimal action point coordinate information of the multi-robot are obtained.
9. The multi-target optimization irregular object-based multi-robot collaborative handling method according to claim 1, wherein after the optimal action point coordinates are obtained, the tail ends of the robot connecting rods are rigidly connected to the object, the formation gesture is adjusted in real time through a controller with rigid formation hard constraints, and the multi-robots are controlled to complete the irregular object collaborative handling task.
10. Multi-robot collaborative handling system based on multi-objective optimization irregular object, characterized by comprising:
The edge contour coordinate extraction module is configured to acquire shape information of an object and extract edge contour coordinates of the object;
the parameter equation construction module is configured to convert edge contour coordinates of an object into a parameter equation of an actual action point of the tail end of the robot connecting rod;
The reachable position point set generating module is configured to expand the shape of the object, establish a mapping relation between a parameter equation and the reachable position and obtain an actual reachable position point set of the robot;
The multi-objective optimization module is configured to construct a multi-objective optimization function, take coordinate points as variables to be optimized, optimize the initial population quality by utilizing an improved NSGA-II algorithm, define constraint violation degrees, further obtain an optimal solution, and determine the coordinates of the optimal action points according to the number of robots;
and the task execution module is configured to control the multiple robots to carry out rigid formation to complete the cooperative carrying task of the irregular object after the optimal action point coordinates are obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510550205.5A CN120066057B (en) | 2025-04-29 | 2025-04-29 | Multi-robot collaborative handling method and system based on multi-objective optimization irregular object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510550205.5A CN120066057B (en) | 2025-04-29 | 2025-04-29 | Multi-robot collaborative handling method and system based on multi-objective optimization irregular object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120066057A CN120066057A (en) | 2025-05-30 |
| CN120066057B true CN120066057B (en) | 2025-06-27 |
Family
ID=95804697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510550205.5A Active CN120066057B (en) | 2025-04-29 | 2025-04-29 | Multi-robot collaborative handling method and system based on multi-objective optimization irregular object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120066057B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11241791B1 (en) * | 2018-04-17 | 2022-02-08 | AI Incorporated | Method for tracking movement of a mobile robotic device |
| CN118863191A (en) * | 2024-06-26 | 2024-10-29 | 同济大学 | AMR collaborative handling method, equipment, and storage medium based on multi-target collaboration |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112621754A (en) * | 2020-12-11 | 2021-04-09 | 中国科学院沈阳计算技术研究所有限公司 | Design method for multi-robot-cooperated assembly line safety layout |
| CN117841006B (en) * | 2024-03-06 | 2024-05-28 | 中建三局集团有限公司 | Track optimization method and device for multiple optimization targets of grabbing manipulator |
-
2025
- 2025-04-29 CN CN202510550205.5A patent/CN120066057B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11241791B1 (en) * | 2018-04-17 | 2022-02-08 | AI Incorporated | Method for tracking movement of a mobile robotic device |
| CN118863191A (en) * | 2024-06-26 | 2024-10-29 | 同济大学 | AMR collaborative handling method, equipment, and storage medium based on multi-target collaboration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120066057A (en) | 2025-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112432648A (en) | Real-time planning method for safe motion trail of mobile robot | |
| CN113296514B (en) | Local path optimization method and system based on sparse banded structure | |
| CN119091433B (en) | Radar-based danger alarm method, device, equipment and storage medium | |
| CN115423972A (en) | A 3D reconstruction method for closed scenes based on vehicle-mounted multi-lidar fusion | |
| JP7574799B2 (en) | Information processing device, information processing method, and program | |
| CN113077484B (en) | Image instance segmentation method | |
| CN114353779B (en) | Method for rapidly updating robot local cost map by adopting point cloud projection | |
| CN120370998A (en) | Unmanned aerial vehicle obstacle avoidance control method and system based on computer vision | |
| CN117409166A (en) | Dynamic object filtering method in 3D point cloud map based on Livox lidar | |
| CN103679814A (en) | Processing method and device for triangle mesh model | |
| CN120066057B (en) | Multi-robot collaborative handling method and system based on multi-objective optimization irregular object | |
| CN116429145B (en) | Automatic docking navigation method and system for unmanned vehicle and garbage can under complex scene | |
| CN113012191A (en) | Laser mileage calculation method based on point cloud multi-view projection graph | |
| CN114419567B (en) | Laser radar target detection method and device based on deep learning | |
| CN119338873B (en) | A 3D point cloud registration method based on multiple convolutional networks | |
| CN114967748A (en) | A UAV Path Planning Method Based on Space Deformation | |
| CN117826129B (en) | A method for on-orbit extrinsic calibration of monocular camera and lidar | |
| CN118447090A (en) | A dynamic visual SLAM method | |
| CN118429416A (en) | A posture optimization method, computer-readable storage medium and intelligent device | |
| Dong et al. | An efficient ground segmentation approach for LiDAR point cloud utilizing adjacent grids | |
| Horng et al. | Vehicle path planning by using adaptive constrained distance transformation | |
| a'Xing et al. | Vision‐based map building and path planning method in unmanned air/ground vehicle cooperative systems | |
| Zhang et al. | Research on the prediction of the operational risk field of intelligent vehicles based on dual multiline LiDAR | |
| CN120068952B (en) | Robot observation data complement model training method and device | |
| Gan et al. | Building feature extraction based on natural neighborhood decomposable point feature extraction algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |